Re: Issues with rif:iri in a number of places in the spec

On Thu, 26 Mar 2009 15:55:59 +0100
Jos de Bruijn <debruijn@inf.unibz.it> wrote:

> > Using xsd:anyURI possibly makes sense for imports.  
> 
> Possibly, but I'm not convinced.  Why not simply write IRIs?  Why use
> constants here at all?

IRI strings would work in Import() in BLD, but in the bigger context they won't.
I explained in a previous email that imported documents may not necessarily have
a location that is known to the authors of the host document (which does
the import).

There is another small technical detail: Hassan noted many times that we need
some kind of a delimiter around the IRI strings used in Base and Prefix. If we
use such strings in Import, then we would need delimiters there as well.
For instance, if we use "...", then effectively we are saying that these things
are xsd:string's that look like IRIs. So, why not make them what they should
really be: xs:anyURIs?


> > If you also wanted metadata that refers to documents, such as a place
> > where you can download the original rule source then that would indeed
> > be an anyURI, for example:
> > 
> >    http://example.com/rule1
> >           dc:author "Dave Reynolds";
> >           eg:originalSource
> >                 "http://dave.reynolds.net/myRules/rule1"^^xsd:anyURI .
> > 
> > Of course annotations have no semantic interpretation within RIF so we
> > can do what we want but this use of rif:iri seems reasonable.  
> 
> Agreed.
> But I think I would also not be opposed to simply writing IRIs here
> (rather than IRI constants). I do not have a preference either way.

The point of making annotations into formulas is to enable reasoning about
them. The ids were defined as constants so that one could write formulas about
those ids (which represent documents) and in this way to exchange
meta-information among reasoners.

An IRI (as a sequence of chars) is not  constant and so you cannot write any
logical statements about it.

michael

Received on Thursday, 26 March 2009 20:22:30 UTC